JPH0375682A - High frequency signal detecting apparatus - Google Patents

Abstract

PURPOSE: To generate a three-dimensional graphical display of digitized geographical data in real time by scanning the correspondent parts of 1st and 2nd data layers from a memory and shading the area which is specified through a process such as removing a hidden face from the scanned parts. CONSTITUTION: This method has a process 10 for scanning the correspondent parts of 1st and 2nd data layers from the memory, process 20 for removing the hidden face from the scanned parts, and process 30 for dividing the scanned parts into the plural cells of 2×2 picture elements so as to form the cell while using the adjacent picture elements from the 1st data layer. Further, this method has a process 40 for further dividing the cell into triangular parts having apexes, process 50 for projecting the apexes on an observation screen while using the coordinate and known gap value of the 2nd data layer, and process 60 for shading the area of which the boundary is specified by the projected apexes. Thus, a realistic photographic image can be generated at high speed from a digitized look-down area photograph combined and aligned with digital geographical contour data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates generally to graphical display systems, and more particularly, to graphical systems such as digital map systems or partial task training devices for flight training, which utilize top-down sensor imaging power and the like. This invention relates to a method for rapidly generating photorealistic images in real time.

[Problems to be solved by conventional technology and invention]

Currently, it is desired to display topographic data or aeronautical chart data on a graphics display terminal for both certain types of aircraft and aircraft training simulators. Such displays may include digital maps of aeronautical charts, radar display images, or other types of images based on photographs or other digitized data.

For example, in the area of digital maps, it is desirable to have the ability to display perspective views of the terrain! Yes. The more realistic the information displayed, the better the performance of the aircraft crew. For example, in an aircraft environment, by adding a perspective map display to the plan view already provided in a conventional digital map system, the crew's ability to judge the situation will be improved. As an example, the transferee's May 1, 1988
Co-pending Application No. 07/192,798 filed on the 1st, Title r DIGITAL MAP SYSTEM J
checking ...

Similarly, the addition of three-dimensional display capabilities will improve the performance of systems such as side surveillance wide area radar systems and aircraft training equipment. In such a system, the display can be performed in real time (i.e. at a rate of 30 images per second).
It is important to display

An example of the results obtained by the present invention is shown in FIG.

FIG. 11a is a typical top-down photograph which, when combined with DMA elevation data using the method of the present invention, yields
A surprising side view is obtained, as shown as a typical example in FIG.

[Means for solving problems]

A method is provided for performing fast generation of photorealistic images for use in a display system.

The display system includes a memory storing first and second suitably modified data layers, the first data layer comprising:
The second data layer is comprised of digitized photographs having known resolution and spacing values, and the second data layer is comprised of digitized elevation data corresponding to the first data layer photographs. The method of the present invention includes the steps of scanning corresponding portions of the first and second data layers from memory, removing hidden surfaces from the scanned portions, and using adjacent pixels from the first data layer. dividing the scanned portion into a plurality of 2 pixel x 2 pixel cells to form a cell, further dividing the cell into triangular parts having vertices, and determining the coordinates of the second data layer. and shading the region bounded by the projected vertices.

In one embodiment of the invention, hidden surfaces are removed by applying a modified Painter's algorithm. Another embodiment of the invention would use two banfas to perform hidden surface removal.

One object of the present invention is to provide a method for rapidly generating photorealistic images from digitized top-down area photographs that are combined and registered with digital terrain elevation data.

Another object of the invention is to generate a side view of the original top-down photograph from any three-dimensional perspective.

Yet another object of the invention is to provide a computer algorithm that generates a three-dimensional graphical representation of digitized terrain data in real time.

Other objects, features, and advantages of the invention will become apparent to those skilled in the art from the description of the preferred embodiment, the claims, and the accompanying drawings. In the drawings, like reference numbers refer to like elements.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

This will be explained with reference to FIG. FIG. 1 shows a flowchart of a method for rapidly generating a database of side view images in accordance with the present invention.

The method of the present invention includes the steps of scanning the memory 10, removing hidden surfaces 20, dividing the layer into cells 30, subdividing the cells into triangles 40, and projecting the vertices 50. and a step 60 of filling the triangle. Briefly, the computer algorithm of the present invention, shown in flowchart form in FIG. 1, receives as input a digitized top-down aerial photograph combined and overlaid with DMA terrain data as shown in FIG.

The computer algorithm of the present invention generates a side view of the original top-down photograph from any three-dimensional perspective. R.D.
An embodiment of the computer algorithm of the present invention, referred to as the Display Algorithm (1), has been implemented in the C programming language on a Silicon Graphics workstation by Honeywell Inc., Minneapolis, Minnesota.

Embodiments of the invention, in the context in which it is implemented, may
Each X 512 pixel database diagram can be generated in 5 seconds. By implementing the RDG-display algorithm on an ultra-high-speed integrated circuit (VI (SIC)), it is thought that the same operation can be performed in real time.An example of such an integrated circuit is the Honeywel+ I
No. 07/3073,548 filed February 7, 1989, filed February 7, 1989 by Assignee, which is also incorporated herein by reference. As described in Mill@r et al. According to FIG. 2, the computer algorithm of the method of the invention assumes that two rectangular layers of co-located data reside in memory. The first layer is the digitized photo 7
It is 0.

The resolution of a photograph is known and is expressed as the constant spacing between successive pixels in the image. Each element in the second layer 72 is a ground elevation corresponding to a pixel at a common location in the photographic layer. In a preferred embodiment of the invention, the resolution of the elevation layer 72 should be equal to the resolution of the photographic layer 700. That is, the spacing between the photographic layers must be equal to the spacing between the elevation layers. The two layers seem to have been corrected correctly when considered as a moat. Essentially, these two layers, a first layer of photographic data and a second layer of elevation data, constitute a three-dimensional photographic representation of a small portion of the Earth's surface.

As those skilled in the art will know, the elevation data required for this computer algorithm is already collected for most areas of the United States, such as
nal Cartographic Informat
It can be purchased from ionCenter (NCIC), etc. High-resolution elevation data is available for Europe and other countries as well. The method of the present invention is N(J
It is of course possible to retrieve the data from another data source, such as binocular photography, rather than relying on elevation data derived from C.

In order to more clearly explain the method of the present invention, the present invention will be described using specific examples. It will be apparent to those skilled in the art that the specific examples used in this example are given for illustrative purposes only and are not intended to limit the invention. In the specific example to be adopted below, the following criteria apply: 1. Elevation data is expressed as a 16-bit integer.

2. The pixel intensity value of the photographic layer is an 8-bit integer.

3. Due to timing reasons, the size of each layer is 512 x 51.
Assume that it has a length of 2 pixels.

4. The ground coordinate system is clockwise 9.

5. The origin of the viewing screen coordinate system is at the lower left corner.

FIG. 3 shows the alignment of the DMA elevation segments to the coordinate system used in discussing the method of the present invention. Reference will now be made to FIG. In Figure 4, 2 pixels ×
Two alternating triangular shapes of two-pixel cells are shown. Each cell 80 is formed into four adjacent pixels 82. Since the four corner points of each cell do not need to lie in the same plane, each cell can be divided into two alternating triangular shapes as shown at 8OA and 80B. The vertices of each triangle are first projected onto the viewing screen using the elevation layer coordinates and spacing values.

Then, the region bounded by the projected vertices is
It is filled using the interpolated part of the Gouraud shading algorithm and the corresponding intensity values taken from the photographic layer. Repeating this "project-fill" operation for all the triangles formed by the 512 x 512 pixels of the two layers results in the striking side view shown in FIG.

Having briefly explained the outline of the method of the present invention above, the process of the present invention will now be explained in more detail. At a certain point, after the memory has been scanned for data to be displayed on the screen, the first processing step consists of removing hidden surfaces. In another embodiment of the invention, two buffers can be implemented in hardware. As those skilled in the art will appreciate, the 2 buffer is a register that stores 2 depth values corresponding to the terrain information being processed for display on the screen. As is well known in the art, two-buffers occur when, for example, some information is located behind another data that is to be displayed at the same time, so that it is "hidden" by that other data. can be used to suppress the writing of that information to screen or frame memory. Here, in order to perform the hidden surface removal function of the present invention, we will use the conventional
Suffice it to say that Banhua can be used.

In another embodiment of the invention, hidden surface removal is performed by using a modified Bainter's algorithm. Hang) rvell Inc, Yo, 6c
The demonstration software implemented in the language version makes very effective use of this method. Since the first and second layers are already spatially organized as plays, no sorting is required, unlike when employing the standard Bainter's algorithm, which requires sorting.

Considering a rectangular photographic layer and an elevation layer, the Painter's algorithm operates by first drawing the surface furthest from the viewer using known rendering techniques.

Delineation continues toward the side closest to the viewer. Essentially, the scene is constructed in a "back to front" fashion.

In this method, hidden surfaces are also depicted, but when a surface closer to the viewer is depicted superimposed on the hidden surface, the hidden surface is covered up.

Next, FIG. 3 will be explained. Here, the three-dimensional position of the observer is assumed to be ■, and the three-dimensional point that the observer is looking at is assumed to be P. Vector difference is equal to p−v. Vector D is D
It points to the observation direction represented by the polar angle phi measured from the X component and the two components.

pht -m-1(Px-Vx)/(P2-V2)MQ
It is realized by setting up four cases depending on the values in the parentheses of the d360 modified Bainter's algorithm H1phi. In the case of one, as shown in FIG. 5, it is realized for each side of a rectangular cell defining four zones.

Range of phl Direction to start drawing 2 315
＜=phi 45 east 1 45
<-ph+135 North 3 13
5<=ph+225 West 0225<=
To render the rectangular layer of the phi315 south zone data, one of four routines is called, depending on the value of phi. As can be seen from FIG. 7, in order to satisfy the first rule of the farthest distance in the Painter's algorithm, it is sufficient to consider two types of triangle orientations. Four routines are required to render successive cells in the correct order, one for each direction. Apart from the directional differences already mentioned, the main difference between each of the four routines is the different order of cell processing. FIG. 9 shows the processing direction of the modified Bainter's algorithm for each zone.

The explanation will be returned to FIG. 6. FIG. 6 shows a view looking down on the rectangular terrain database.

For convenience in describing the Reconsolidation Painters algorithm of the present invention, it is assumed that the terrain 100 is flat, but the following discussion applies to terrain databases of arbitrary complexity. The line of sight vector shown in the lower area of the database points in the right direction. The small square is 2x2 pixel cell 8
Represents 0. The database contains the following information for each vertex B2 of each 2x2 cell 80 in the database: 1. Based on a priori known spacing parameter (x,
y coordinate 2, representing the a-height that is removed from the elevation layer 2 coordinate 3, the intensity value extracted from the photographic layer When using the modified Bainter's algorithm of the present invention, first, the line of sight vector D is , for example, as shown in FIG. 5, move to the center of the database and determine the zone containing the moved line-of-sight vector using the angle phi. Next, as shown in FIG. 9, an appropriate zone-related rendering procedure is selected.

Next, the appropriate topographical map is drawn one cell at a time.

As shown in FIG. 7, each cell is drawn by first decomposing the cell drawing operation into two triangle drawing operations. After the appropriate rendering procedure has been selected, there are two possible ways to render the cells. The first method uses a diagonal line that crosses the cell from the top left to the bottom right as shown in FIG.
A diagonal line that crosses the cell from the bottom left to the top right is used as shown in Figure (b).

Within a single cell, experiments have shown that either diagonal can be chosen for all cells in all zones without significant effects. What is important in all zones is that the two triangles of a cell are drawn in the correct order. In either case, a triangle located far away from the line-of-sight vector D must be drawn first. Furthermore, the row of cells furthest from the viewer must be rendered before the row of cells closest to the viewer. Also, the diagonal M selected for all zones must be consistent to avoid artifacts during real-time image generation.

One of the major advantages of this method is that there is no need to sort the database to determine which cells to render. This is because the data and photographic layers are stored in computer memory in a spatially ordered manner.

All four cases, i.e., all four zones, use a memory that can scan successive lines of cells in a forward or reverse direction, or scan columns of cells in a forward or reverse direction. It would be good if it was depicted as such. This four-mode memory read capability corresponds to the four zones described above and is implemented in demonstration software.

Assume that the simulated vehicle is flying through the terrain database. As the pht changes and the photographic image is rendered, a transition occurs somewhere between the four rendering procedures. As long as the diagonals selected for all cells in all zones are consistent, the transitions between the rendering steps corresponding to each zone will not generate artifacts, i.e., the transition from zone 1 to zone 2 will not produce artifacts. It has been demonstrated that the sequence of scenes generated is an undetectable motion that is invisible to the human eye.

This quality is desirable and makes the method described herein highly suitable for flight simulator applications.

Consider one of the four boundary conditions shown in FIG.

FIG. 8 shows the database shown in FIG. 6 from a viewpoint along the line-of-sight vector. In this case, the line of sight vector D locates zone 1 and zone 2 exactly along the joining line. In this case, the rendering procedure can be selected from either Zone 1 or Zone 2 without producing artifacts that appear confusing to the observer.

Now that the hidden surface removal method employed in the computer algorithm of the method of the present invention has been explained, the graphic display operation required for each triangle will now be explained. The operations required for each triangle are as follows: 1. Project the three vertices of each triangle onto the screen.

2. Fill the projected triangles using interpolative shading, ie, modified Gourand shading.

For the first step of projecting the vertices onto the screen, we have to multiply each 3D triangle vertex by the perspective transformation matrix. The contents of the transformation matrix itself are determined uniquely on a frame-by-frame basis, and the use of such transformation matrices is well known in the art. Taking into account all rotations, translations, and transformations from the observer coordinate system and projection onto the viewing screen, the final matrix has the following form:

rll r12 r13
0r21 r22 r23
0r31 r32 r33
Since the r term is the product of sine and cosine, it should lie between the values 0 and 1. In this example, for convenience, the matrix coefficients are all 32-pict floating point values. 512
Taking a layer of X 512 pixels as an example, 26.2144 matrix multiplication operations must be performed to project all the triangles onto the screen.

Once the vertices of the triangle have been projected onto the viewing screen, a filling operation is now performed according to the method of the invention. After being projected onto the screen, it becomes a two-dimensional triangle. This triangle is then filled using a modified Gouraud shading algorithm.

In its purest form, Gouraud shading is
The first three processes are attributed to the strength of each vertex. When applying the well-known Gouraud shading to the three-dimensional image to be produced in accordance with the method of the invention, the intensities are obtained directly from the photographic layer and are thus known values. In another embodiment of the invention, a lighting (xtghttng) model may be added.

The lighting model essentially consists of firstly determining the surface normal (cross product) and secondly determining the illumination intensity (point product) from the surface normal and light vector. Conceivable.

The illumination value is then decomposed into intensity and number of times during the filling operation.

This operation must be calculated once for each triangle.

As a result, the remaining operation is a triangle filling operation, which is essentially a bilinear interpolation. The interpolation occurs along the sides of the triangle, without interpolation, and then between the sides along each scan line of the screen.

ll 78 'l '8 The above formula is
Represents the value for strength interpolation along the sides of the triangle. The triangle has apex 11. It has a known strength in I2 and Technique 3. I
, , I, and I are the intensities along the scan line S and are interpolated from II, I2, and I3 according to the equations above. This is a well-known method employed in the prior art.

The foregoing has described the present invention in a manner consistent with patent law, and provides information necessary for those skilled in the art to apply the novel principles and to construct and use such specialized components as desired. As if to give, I explained in detail. It is understood, however, that the invention may be carried out with essentially different equipment and apparatus and that various modifications may be made both as to details of the apparatus and as to operating procedures, without departing from the scope of the invention itself. I have to do it.

[Brief explanation of drawings]

FIG. 1 is a flowchart illustrating the method of the present invention for rapid database generation of side view images, and FIG. 2 shows DMA elevation information used to generate side view images according to the method of the present invention. FIG. 3 is a diagrammatic representation of a photographic layer and an elevational layer as shown in FIG. 2 to be utilized by the method of the invention. I) A graph showing the coordinate system, FIG.
FIG. FIG. 6 shows the partitioning of the space into four zones based on the polar angle phi extracted from the parent line vector D used in the modified Bainter's algorithm adopted in the method of the invention, FIG. 7 is a view looking down on a rectangular terrain database containing small cells representing 2×2 pixel cells, with a gaze vector pointing to the right located in the lower area of the database. FIG. FIG. 8 shows the order of processing a 2×2 pixel cell using Bainter's algorithm; FIG. 9 is a diagram illustrating the relationship between the processing direction of the Painter's algorithm and the four zones adopted in the method of the invention; FIG. Figure 11a is a top-down view of a three-dimensional representation of the kind obtained according to the method of the invention using the image of Figure 11a. It is a figure showing one sample. 80...Cell, 80A, 80B...Triangle, 82...Vertex.

Claims (1)

[Claims] (1) a memory storing first and second suitably modified data layers, the first data layer consisting of a digitized photograph having known resolution and spacing values; provides a method for performing rapid generation of photorealistic images for use in a digital display system, comprising digitized elevation data corresponding to photographs in a first data layer, comprising: (a) a first data layer; (b) removing hidden surfaces from the scanned portion; (c) resolving cells using adjacent pixels with respect to the first data layer; dividing the scanned portion into a plurality of 2 pixel by 2 pixel cells so as to form; (d) further dividing the cell into triangular portions having vertices; (e) a second data layer; (f) sizing the region bounded by the projected vertices.